JP2002055985A - Device and method for document attribute discrimination - Google Patents

Abstract

PROBLEM TO BE SOLVED: To lighten the burden of document attribute adding operation on a user by making it possible to discriminate document attributes written on a paper document itself when document management is performed by using a computer. SOLUTION: A document managing server 3 is mounted with a document attribute extraction information generating AP 37, an area information DB 15, and a document management processing AP 10. The user specifies the kinds, entry positions, and entry position IDs of document attributes for an image-documented paper document where the document attributes are described by using the extraction information AP 37 and stores them in the area information DB 15. Further, a document attribute extraction sheet is generated in which the kinds and entry position IDs of the document attributes as objects of extraction are set. When the paper document is scanned, the sheet is scanned as the cover of the paper document. The document management processing AP 10 discriminates and analyzes the scanned sheet, extracts document attributed from the paper document scanned after the sheet on the basis of the kinds and entry position IDs of the document attributes set on the sheet and the information in the area information DB 15, and gives them to an image document of the paper document.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique relating to document management using a computer, and more particularly, to the attributes of an image document created by reading a paper document such as a proposal, a meeting record, and a payment with a scanner or the like. Information (document name, author,
The present invention relates to a technology for identifying a document attribute (hereinafter referred to as a document date and time).

[0002]

2. Description of the Related Art In various institutions such as banks and companies, generally, paper documents such as proposals, minutes, and settlement documents are read by a scanner or the like to form a scan image (hereinafter, referred to as an image document). Input to the computer and management are performed. At this time, document attributes such as a document name, a creator, and a creation date and time are assigned to each image document.

As a method of assigning a document attribute to an image document, an operator can directly assign the attribute based on a document attribute described in a paper document,
(Optical Character Reader) Prepare a paper for assigning document attributes called a form, and manually enter the prescribed document attributes in a prescribed position on the document attribute assignment paper (printing is possible if it can be entered in the prescribed position). There is a method in which the filled-in sheet is provided as a cover of a paper document and read by a scanner or the like together with the paper document. In the latter case, the OCR software (application software having an OCR function) mounted on a computer or a scanner regards the first sheet of the paper document as a document attribute-added sheet, and writes the entry contents from each predetermined position on the paper. The content is extracted and recognized, and the recognized entry content is added to the image document as a document attribute.

[0004]

Various paper documents such as proposals, minutes, and settlement documents include document attributes such as a document name, a creator, and a creation date and time. Document attributes written on the document itself cannot be identified. Therefore, when assigning document attributes to an image document, as described above, the user directly assigns the document attributes to the computer by operating keys, or writes the document attributes by handwriting or printing on a document attribute assignment sheet. Cumbersome work of granting and giving.
In any of these operations, generally all characters related to the document attribute must be entered. Therefore, when the number of characters in the document attribute is large (for example, when the document name "Revision of a Cabinet Order concerning International Registration Application and International Trademark Registration Application and Practical Operation This is particularly troublesome when these characters in "about" must be entered).

In addition, this operation needs to be performed every time one paper document is converted into an image document.

As described above, in document management using a computer, the task of assigning document attributes is cumbersome and burdensome for the user.

Accordingly, an object of the present invention is to make it possible to identify a document attribute written in a paper document itself in document management using a computer, thereby reducing the burden on the user for assigning document attributes. It is in.

[0008]

SUMMARY OF THE INVENTION A document attribute identification device according to the present invention indicates position information indicating a position of a document attribute described in a paper document to be an image document, and indicates a type of the document attribute described in the position. Input means for inputting document attribute-related information including type information; storage means for storing the input document attribute-related information; and document attributes of a desired type among document attributes described in a paper document to be an image document Specifying means for specifying the position information of the document, identifying means for extracting a document attribute from the image document of the paper document based on the specified position information and the stored document attribute related information, and identifying a type of the extracted document attribute. Is provided.

According to the present invention, it is possible to extract a user-desired document attribute from the paper document itself and identify the type of the document attribute only by specifying the position information of the desired document attribute in the paper document. The extracted document attribute can be added to the image document related to the document attribute and stored together with the image document.

In a preferred embodiment, the document attribute-related information includes a position identifier assigned to the position information, and the specifying means specifies the position identifier of the desired type of document attribute, and Extracts the document attribute from the image document of the paper document based on the designated position identifier and the document attribute related information, and identifies the type of the extracted document attribute.

In a preferred embodiment, the document attribute-related information includes character feature information indicating a feature of a character described in the document attribute, and the designating means determines the character feature information of the character described in the desired type of document attribute. And identifying means for extracting a document attribute from the image document of the paper document based on the specified character characteristic information and the document attribute related information, and identifying the type of the extracted document attribute.

[0012]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing the overall configuration of a system according to one embodiment of the present invention.

This system scans various paper documents such as conference records and settlement documents to create image documents, and scanners 1a, 1b,. Document management server 3 that extracts and assigns it to the image document to manage the document
And The scanners 1a, 1b,... Are used, for example, as peripheral devices of the document management server 3.

FIG. 2 is a block diagram showing the configuration of the document management server 3.

The document management server 3 is, for example, a general-purpose computer such as a personal computer or a workstation, and includes a display 31, an input device 33 such as a keyboard and a mouse, and the image document itself created by the scanners 1a, 1b,. Application software (hereinafter referred to as an extraction information creation AP) 37 for creating a document attribute extraction sheet for setting information for extracting a document attribute from the document.
Further, the document management server 3 stores a document attribute extraction sheet format database (hereinafter, sheet format DB).
11 and area information database (area information DB) 1
5, a character information database (character information DB) 19,
A document attribute / image document database (document attribute / image document DB) 27 and a document management processing application (document management processing AP) 10 are provided. Document management processing A
P10 recognizes the setting content of the document attribute extraction sheet, extracts the document attribute from the image document itself according to the setting content, and attaches it to the image document to manage the document.
More specifically, the document management process AP10 includes a scanner control unit 7, a document attribute extraction sheet identification unit (hereinafter, sheet identification unit) 9, a document attribute extraction sheet analysis unit (hereinafter, sheet analysis unit) 13, Area information analyzing section 17, character information analyzing section 21, document attribute extracting section 23, document attribute assigning section 2
5 is provided.

The extraction information creation AP 37 includes a scanner 1a,
1b,..., An image document in which the document attributes are described, is received and displayed on the display 31, and the description positions (page numbers and description positions in the pages) of various document attributes described in the displayed image document are received. The user is caused to create a document attribute extraction sheet, specify a document attribute area, and set information on the document attribute extraction sheet based on the characteristics of the document and the description. In the document attribute extraction sheet, area information indicating the writing position of the document attribute, character feature information (font, character size, and the like) of the document attribute are set (details will be described later).

In the sheet format DB 11, information on a document attribute extraction sheet created by the user is stored by an extraction information creation AP 37.

In the area information DB 15, information relating to the document attribute area designated by the user is stored by the extraction information creation AP 37.

The character information DB 19 contains a document management process AP.
A font / font ID table, which will be described later, for specifying a character font having a document attribute and referred to by the character information analysis unit 21 is stored.

The document attribute / image document DB 27 stores an image document to which a document attribute has been assigned by the document attribute assigning unit 25 of the document management process AP10.

The scanner control unit 7 includes scanners 1a, 1
b,... are periodically monitored to determine whether a paper document is being scanned. If there is a scanner that is scanning the paper document, an image document is acquired from the scanner.

The sheet identification unit 9 checks each page of the acquired image document and determines whether or not each page is a document attribute extraction sheet. When identifying the document attribute extraction sheet, the sheet identification unit 9 notifies the sheet analysis unit 13 of the identification.

The sheet analysis unit 13 acquires various information set by the user from the identified document attribute extraction sheet based on the information stored in the sheet format DB 11.

The area information analysis unit 17 converts the image document itself based on the document attribute description position information of the various information obtained from the document attribute extraction sheet and the area information stored in the area information DB 15. The description area of the described various document attributes is specified, and information of the specified area is notified to the document attribute extraction unit 23.

The character information analysis unit 21 is based on character characteristic information (for example, font) of various information obtained from the document attribute extraction sheet and a font / font ID table stored in the character information DB 19. The character attribute of various document attributes described in the image document itself is specified, and information of the specified character characteristic is transmitted to the document attribute extracting unit 2.
Notify 3.

The document attribute extracting unit 23 analyzes the image document, and extracts the document attributes from the image document itself based on the information notified from the area information analyzing unit 17 and the information notified from the character information analyzing unit 21. Extract.

The document attribute assigning section 25 includes the document attribute extracting section 2
3 assigns the extracted document attributes to the image document, and stores the image document to which the document attributes have been assigned in the document attribute / image document DB 27.

Hereinafter, the processing performed by the document management server 3 will be specifically described, taking as an example the case where the image document in which the document attributes are described is the image document 41 shown in FIG.

First, the creation of the document attribute extraction sheet, the division of the document attribute area, and the setting of information on the document attribute extraction sheet will be specifically described.

The image document 41 shown in FIG. 3 is, for example, the first page of an image document having a plurality of pages, in which various document attributes are described. For example, as the types of document attributes, version information “ Version 1.0 ”, document name “Proposal on XX”, company / organization name “
○ ”, author“ ○ × Taro ”, creation date“ 2000/01/01 ”
(Of course, various document attributes are described not only on the first page but also on a plurality of pages in some cases). The user creates the following document attribute extraction sheet using the extraction information creation AP 37 based on these document attributes.

FIG. 4 shows an example of a document attribute extraction sheet.

The document attribute extraction sheet 43 is a paper document itself.
(That is, extraction described in the image document 41 itself)
Description of target document attributes (document name, creator, etc.) and document attributes
Placement position (page number and position on the page
) And document attributes (character size, phone
Sheet, etc.) can be set. More specifically
In other words, the document attribute extraction sheet 43 includes the document attribute extraction system.
Unique ID (sheet ID)
Entry field 45 and items that can be set for various document attributes
Attribute item expansion column 49 in which (hereinafter, attribute items) are expanded.
And a document attribute input field 4 for inputting a document attribute to be extracted
7 and information about each attribute item of the input document attribute.
Information setting box 51 for specifying₁~ 51_n(In the figure
51₁~ 51 ₃₀) Are provided.

In the attribute item expansion column 49, recordable attribute items for various document attributes are arranged in the horizontal direction. Types of the attribute items to be arranged include, for example, “description page”, “area”, “font”, “size”, “character decoration”, and “number of acquired characters”.
The “description page” is an attribute item for setting a page number in which a document attribute to be extracted is described. “Area” is an attribute item for setting in which area of the page set in “Description Page” the document attribute to be extracted is described. “Font” is an attribute item for setting the font of the character of the document attribute to be extracted. “Size” is an attribute item for setting the size of the character of the document attribute to be extracted. "Character decoration" means what character decoration (italics, italic,
This is an attribute item for setting whether or not the item is underlined. “Acquired number of characters” is an attribute item for setting the number of characters to be extracted. The attribute item may be arbitrarily added by the user.

The document attribute input column 47 arranges document attributes to be extracted in the vertical direction. In the input field 47, the user can extract the types of document attributes described in the image document 41, such as “document name”, “creator”, “company / organization name”, and “year of creation”. Enter the “month and date” and “version”. Note that some document attributes to be extracted may be prepared by default.
In that case, the user can arbitrarily add / delete.

The information setting box 51 _{1-51 30} includes a respective attribute items developed in the attribute item expand column 49, is deployed on the intersection of the respective document attributes entered in the document attribute input column 47. The user, each information setting box 51 _{1-51 30,} in accordance with the attribute item information setting target, so as to set either check marks or numbers. This setting is performed by directly setting data on the document extraction and creation sheet 43 by using the client machine 5a, or by printing out the document extraction and creation data 43 by handwriting.

As described above, the user creates the document attribute extraction sheet 43 using the extraction information creation AP 37. The extraction information creation AP 37 analyzes the created document attribute extraction sheet 43 and acquires the following information. That is, the extracted information creation AP37 includes position coordinates of each information setting box _{51 1-51 30} when the as a reference (origin) (upper left corner of the example document attribute extraction sheet 43) a predetermined position, each information setting box 51 and attribute items and document attribute type _{information 1-51 30} corresponds, obtains the information of the setting method for each information setting box 51 _{1-51 30} (to set either check mark and numbers). The extraction information creation AP 37 saves the acquired information in the sheet format DB 11.

The user, in the information setting box 51 _{1-51 3 0} of the document attribute extraction sheet 43, based on the features described positions and descriptions of various document attributes that are listed in the image document 41, each attribute Set information about the item. Hereinafter, a method of setting information for each attribute item will be described.

[0039] information setting box 51 _{1-51 5} attribute item "according page" are each unique document attribute sets the number of pages that have been described. In terms of image document 41 shown in FIG. 3, numbers are set in the information setting box 51 _{1-51 5} are all "1".

Information setting box 5 for attribute item “Area”
1 ₆ to 51 ₁₀ sets the area ID assigned in the following manner. Referring to FIG. 5 and FIG. 6, the area ID
Will be described.

FIG. 5 shows a display screen of the image document 41 when the description area of the document attribute is designated.

The user operates the mouse to surround various document attributes on the image document 41 displayed on the display 31 with a frame of an arbitrary shape (square, ellipse, etc.) and set an area in the image document 41. For example, as shown in FIG. 5, the user adds an area 61 enclosing version information “ Version 1.0 ” with a rectangular frame to the image document 41,
Area 63 enclosing the document name “Proposal on XX” in a square frame, area 65 enclosing the company / organization name “XX Corporation” in a square frame, and the creator “XX Taro” in a square frame And an area 69 enclosing the creation date "2000/01/01" in a rectangular frame. After the user designates the areas 61 to 69 for the image document 41 as described above, the user designates each area 61 to 69 for each of the areas 61 to 69.
A unique area ID for identifying ６９69 is assigned.

FIG. 6 is a display screen when an area ID is assigned to each of the areas 61 to 69.

After specifying the areas 61 to 69 in the image document 41, the specified areas 61 to 69 are displayed together with the display area 41a of the image document 41. The user assigns a unique area ID (for example, a number) to each of the areas 61 to 69 on this screen. For example, the user sets area ID “1” in area 61 and area 6
3 is the area ID "2", area 65 is the area ID
"3", area 67 has area ID "4", area 69
Is assigned an area ID "5".

The extraction information creation AP 37 analyzes the image document area 41a to which the area ID is assigned, obtains each area ID and the position information of the area to which each area ID is assigned, and stores the information in the area information DB15. To save. The location information of the area is stored in the image document area 4
The position when a predetermined position 1a (for example, the upper left vertex) is set as a reference (origin), for example, the coordinates of each vertex if the shape of the area is a rectangle, and the center coordinates and the information of the equation representing the ellipse if the area is an ellipse is there.

The work of assigning the area ID as described above may be performed before or after the creation of the document attribute extraction sheet.

Referring back to FIG.

Information setting box 5 for attribute item “Area”
1 _6-51 The _10, the area corresponding to the various document attributes I
Set D. That is, according to the above-described area ID assignment operation, the correspondence between the area ID and the document attribute type is as follows.
“ID“ 1 ”:“ version ””, “ID“ 2 ”:“ document name ””, “ID“ 3 ”:“ company / organization name ””, “ID
"4": "Creator,""," ID "5": so has become a "creation date", "the user, the information set box 51 ₆ of the document attribute" document name "," 2 ", the information set box 51 ₇ of the document attribute "creator", "4", the information set box 51 ₈ of the document attribute "companies and organization name", "3", the document attribute "creation date" in the information setting box 5
The 1 ₉ in the information setting box 51 ₁₀ "5", document attribute "version" is set to "1".

[0049] information setting box _{51 11-51 15} property item "Font", based on the font document attributes that are listed in the image document 41, font / font ID previously stored in the character information DB19 Set the information according to the table.

FIG. 7 shows a font / font ID table.

In the font / font ID table 71, a font name and a font ID (for example, a numeral) corresponding to each font name are recorded. For example, the font / font ID table 71 shown in FIG.
In the order of "font name: font ID", "Gothic: 1", "Mincho: 2", "Square: 3", "Thick Gothic: 4", and "Pop: 5" are described. . This font / font ID table 71 may be created by the user.

In accordance with the font / font ID table 71, the user selects the attribute item “Font” shown in FIG.
Setting the information of the information setting box _{51 11-51 15.} The information setting box 5 of the attribute item “Font”
If not set information in 1 _{11-51 15,} extracts the document attribute is performed regardless of the font of the document attributes.

Information setting box 5 for attribute item “size”
1 _16-51 to _20, points to the character size of the document attributes to be extracted (predetermined word processing software (e.g., Micr
Set with the character size that can be set in Word98) of osoft. The information setting box 51 of the attribute item “size”
_{16-51 20} when not set information, the extraction of document attribute is performed regardless of the character size of the document attributes.

[0054] in the attribute item information set box 51 ₂₁ to 51 ₂₅ of "character ornament" is, in the character of the document attributes to be extracted, fill in italics, information set when there is a character ornament underlined, such as, for example, a check mark I do. In the image document 41 shown in FIG. 3, since the document attribute "version" is described as " Version 1.0 ", the character decoration "underlined" is displayed, and the document attribute "creation date" is "2000/01/2000". since it is described that 01 "because of the character embellishment" italic ", the document attribute" version "and" correspondence information set creation date "box 51 ₂₄ and 51 _25, is a check mark.

Of course, in this embodiment, character decoration IDs corresponding to various character decorations are prepared (for example, character decoration ID “1” for character decoration “bold”, character decoration ID “1” for character decoration “Italic”). Prepare a table in which character decoration ID “3” is associated with ID “2” and character decoration “underlined”. If the character of the document attribute to be extracted has character decoration, it corresponds to the character decoration. The entered character decoration ID may be entered. It should be noted that the information set box 51 ₁₆ of the attribute item "character ornament"
If the to 51 ₂₀ no information setting, the extraction of document attributes is carried out regardless of whether or not there or character ornament is in the document attribute.

[0056] The information set box 51 _{26-51 30} of the attribute item "Get the number of characters" is, when you want to limit the number of characters of the document attributes to be extracted, to set the number of characters.

[0057] As described above, the user sets the information for extracting various document attributes from the image document 41 in the information setting box 51 _{1-51 30} document attribute extraction sheet 43, and a sheet ID entry A unique sheet ID is entered in the column 45 to complete the document attribute extraction sheet 43 'shown in FIG.

When the user scans a paper document related to the document attribute extraction sheet 43 'with an arbitrary scanner 1a (or 1b,...), The document is placed at the top of the paper document (that is, as a cover of the paper document). The attribute extraction sheet 43 'is set (that is, the document attribute extraction sheet 43' is first scanned by the scanner 1a). When one scanner 1a is to scan a plurality of types of paper documents, a document attribute extraction sheet is set at each boundary where the types of paper documents are different.

The document attribute extraction sheet 43 'and the paper document set on the scanner 1a are scanned by the scanner 1a. As described above, the document management processing AP 10 of the document management server 3 identifies and analyzes the document attribute extraction sheet 43 ′, and sets the scanned document attribute extraction sheet 43 ′ and the paper document as described above. The attribute information is extracted from the image document created by scanning after the sheet 43 ′ according to the information, and the extracted attribute information is added to the image document.

The processing flow of the document management process AP10 when the paper document to which the above-described document attribute extraction sheet 43 'is attached is scanned by the scanner 1a will be described below with reference to FIG.

One of the scanners, for example, the scanner 1a
If a paper document is being scanned (Yes in step S1), the scan data created by the scanner 1a is acquired (S2). Document management processing AP10
Checks each page of the acquired scan data (S3), and determines whether or not the page is the document attribute extraction sheet 43 '(S4). When the document attribute extraction sheet 43 'is identified (Yes in S4), it is determined that the image document of the previous page and the image document of the subsequent pages are different, and the image document of the pages subsequent to the sheet 43' is determined. Extract document attributes from image documents.

When the sheet 43 'is identified (Y in S4)
es) The document management process AP10 is in a sheet format
Information in DB 11, ie, document attribute extraction sheet 43 '
Information setting box 51 in₁~ 51₃₀Position coordinates
And each information setting box 51 ₁~ 51₃₀Genus corresponding
Type information of gender items and document attributes, and each information setting box
51₁~ 51₃₀Setting method (check mark and number
Sheet 4) based on the information
3 'is analyzed, and each information setting box 51 of the sheet 43' is analyzed.
₁~ 51₃₀The information set in (1) is obtained (S5).

Next, the document management processing AP 10
Wherein the position information of the document attribute acquired from the information setting box _{51 1-51 10} 3 '(page number and area I
Based on D) and the area information (area ID and area position information corresponding to the area ID) stored in the area information DB 15, document attributes of various extraction targets described in the image document itself are described. The area (position) is specified (S6). The document management process AP10 the character characteristic information (font ID of the document attributes acquired from the information setting box _{51 11-51 30} sheets 43 ', the character size point, check mark indicating the presence or absence of character embellishment, and acquires the number of characters ) And the font / font ID table stored in the character information DB 19, the character characteristics of the document attributes of various extraction targets described in the image document itself are specified (S7).

Then, the document management process AP10 analyzes the image document and extracts document attributes from the image document based on the specified description area and character characteristics (S).
8). The document management process AP10 assigns the extracted document attributes to the corresponding image document (that is, the image document of the page from the next page identifying the sheet 43 ′ to the page before identifying another document attribute extracting sheet) ( S9) The image document is stored in the document attribute / image document DB 27 (S10).

As described above, according to the above-described embodiment, the user sets the desired number (page number,
If a sheet ID, area ID, font ID, etc.) or a check mark is entered, and the sheet is used as the cover of the paper document and scanned, the document attributes are extracted from the paper document itself, and the document is added to the image document. Attributes can be assigned. Since the information to be entered by the user in the document attribute extraction sheet 43 is only numbers or check marks, no matter how many characters are included in the document attribute, the amount of entry is small. In other words, the document attribute assigning operation, which was conventionally cumbersome and burdensome, becomes relatively easy for the user.

Further, according to the above-described embodiment, the information to be entered in the document attribute extraction sheet 43 is sufficient only at the position where the document attribute of the paper document to be scanned is written. (That is, if the type of the document attribute described in the paper document and the position of the document attribute are the same), the document attribute extraction sheet 43 in which the information has been once written
Can be copied and reused (previously, even if the scanned paper document was the same format,
Since the content of the document attribute itself is different (for example, the document name itself is different even if the description position of the document name is the same), after all, the document attributes such as the document name and the creator must be input. ). Various paper documents, such as proposals, minutes, and settlement documents, often have the same format if they are the same type of paper document. Therefore, as described above, if the document attribute extraction sheet that has been used once can be copied and reused, it is very convenient and efficient for the user.

Although the preferred embodiment of the present invention has been described above, this is an exemplification for describing the present invention, and is not intended to limit the scope of the present invention only to this embodiment. The present invention can be implemented in other various forms.
For example, "document type" is additionally registered as a document attribute,
It is also possible to manage documents based on document types such as specifications, design documents, proposals, minutes, settlement documents, orders, and papers.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the overall configuration of a system according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a document management server 3.

FIG. 3 is a diagram showing an example of a scanned image document 41.

FIG. 4 is a diagram showing an example of a document attribute extraction sheet.

FIG. 5 is a view showing a display screen of an image document 41 when a description area of a document attribute is designated.

FIG. 6 is a view showing a display screen when an area ID is assigned to each of areas 61 to 69;

FIG. 7 is a view showing a font / font ID table.

FIG. 8 is a view showing a document attribute extraction sheet when information is set based on an image document 41.

FIG. 9 is a diagram showing a processing flow of a document management process AP10 when a paper document attached with a document attribute extraction sheet 43 ′ is scanned by the scanner 1a.

[Explanation of symbols]

1a, 1b,... Scanner 3 Document management server 7 Scanner control unit 9 Document attribute extraction sheet identification unit (sheet identification unit) 11 Document attribute extraction sheet format database (sheet format DB) 10 Document management processing application (document management process A)
P) 13 Document attribute extraction sheet analysis unit (sheet analysis unit) 15 Area information database (area information DB) 17 Area information analysis unit 19 Document information database (character information DB) 21 Character information analysis unit 23 Document attribute extraction unit 25 Document attribute Assignment unit 27 Document attribute / image document database (document attribute / image document DB) 31 Display 33 Input device 37 Document attribute extraction information creation application (extraction information creation AP)

Claims (5)

[Claims] An input for inputting document attribute related information including position information indicating a position of a document attribute described in a paper document to be an image document and type information indicating a type of the document attribute described in the position. Means for storing the input document attribute-related information; specifying means for specifying position information of a desired type of document attribute among document attributes described in a paper document to be the image document; A document attribute identification device comprising: identification means for extracting a document attribute from an image document of the paper document based on designated position information and the stored document attribute related information, and identifying a type of the extracted document attribute. 2. The document attribute-related information includes a position identifier assigned to the position information, wherein the specifying unit specifies a position identifier of position information of the desired type of document attribute, 2. The document attribute identification device according to claim 1, wherein the document attribute identification device extracts a document attribute from the image document of the paper document based on the designated position identifier and the document attribute related information, and identifies a type of the extracted document attribute. . 3. The document attribute-related information includes character feature information indicating a feature of a character described in the document attribute, and the specifying unit specifies the character characteristic information of the character described in the desired type of document attribute. 2. The method according to claim 1, wherein the identifying unit extracts a document attribute from the image document of the paper document based on the designated character feature information and the document attribute related information, and identifies a type of the extracted document attribute. Document attribute identification device described. 4. A step of inputting document attribute related information including position information indicating a position of a document attribute described in a paper document to be an image document and type information indicating a type of the document attribute described in the position. Storing the input document attribute-related information; specifying the position information of a desired type of document attribute among the document attributes described in the paper document to be the image document; Extracting a document attribute from the image document of the paper document based on the position information and the stored document attribute-related information, and identifying a type of the extracted document attribute. 5. A step of inputting document attribute related information including position information indicating a position of a document attribute described in a paper document to be an image document and type information indicating a type of the document attribute described in the position. Storing the input document attribute-related information; specifying the position information of a desired type of document attribute among the document attributes described in the paper document to be the image document; Extracting a document attribute from the image document of the paper document based on the location information and the stored document attribute-related information, and identifying a type of the extracted document attribute. Computer readable recording medium.